CICEET Progress Report for the period 3/15/05 Through 9/15/05

Project Title: Wastewater Treatment to Minimize Nitrogen Delivery from Dairy Farms to Receiving Waters
Principal Investigator(s): Nancy G. Love
Additional Investigator(s): Katharine F. Knowlton, Barth F. Smets
Project Start Date: November, 1 2004 (Delayed due to date of funding arrival).

Figures

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Tables

Table 1

Table 2

Table 3

Objective 2. Experimentally determine the operating parameters and treatment capacity for an OLAND-based treatment system.

Objective 3. Characterize the liquid, solid and gaseous effluent from the OLAND process.

Objective 4. Develop practical design equations and performance curves for the OLAND-based biofilm treatment system.

Objective 5. Ensure environmental compatibility by calculating a nitrogen balance for a model concentrated dairy operation located in the NERR region.

This report addresses specific activities associated with Objectives 1, 2 and 5. Objectives 1 and 2 will be reported simultaneously because the tasks associated with running the OLAND reactor systems overlap significantly.

Tasks to meet objectives
To achieve objective 1, the following subtasks are currently being pursued:

1. Continued operation of two (2) OLAND lab-scale reactors and construction of a third OLAND lab-scale reactor and one OLAND pilot-scale reactor.
2. Construction and continued operation of a pilot plant-scale anaerobic storage tank using traditional design parameters so that a crust forms on the stored waste and holds the ammonia in.
3. Evaluation of the performance of the anaerobic storage tank to stabilize organic matter and generate reduced nitrogen.
4. Evaluation of the ability of the OLAND reactors to treat stabilized dairy waste by:
   1. Establishing stable autotrophic nitrogen removal using synthetic wastewater, and
   2. Demonstrating autotrophic nitrogen removal using anaerobically-stabilized dairy manure.

To achieve objective 2, the following subtasks are/will be completed:

1. Construction of at least one pilot-scale OLAND reactor system for treatment of anaerobically stabilized dairy wastes.
2. Operation and maintenance of lab and pilot-scale OLAND systems utilizing anaerobically stabilized dairy waste.
3. Exposure of said reactor systems to increasingly higher nitrogen loading conditions and evaluation of the performance associated with reactor configurations.
4. Evaluation of optimal reactor configuration and startup procedures for implementation of OLAND systems at both lab and pilot-scale.

To achieve objective 5, the following subtasks are/will be completed:

1. Evaluate the waste composition from three dairy farms located in the Winyah Bay and compare to our pilot-scale anaerobic stabilization reactor to determine which farm is best suited for performing the full-farm nitrogen balance.
2. Complete a nitrogen balance using current operating protocols for the selected dairy farm in the Winyah Bay watershed in South Carolina.
3. Computationally assess the impact of implementing the OLAND technology on nitrogen cycling around the farm by recalculating the nitrogen balance, assuming OLAND is in place.
4. Revise the nitrogen balance calculations at the end of the project, once further information about OLAND performance with anaerobically stabilized dairy waste is completed.

To achieve objective 5, the following subtasks are/will be completed:

1. Evaluate the waste composition from three dairy farms located in the Winyah Bay and compare to our pilot-scale anaerobic stabilization reactor to determine which farm is best suited for performing the full-farm nitrogen balance.
2. Estimate all needed flows and concentrations on the farm to complete the Nitrogen balance.

Progress on Tasks
Fixed bed bioreactor (FFBR) Operation and Maintenance.
Prior to this reporting period, a continuous flow submerged fixed bed bioreactor (FBBR) was constructed and operated in an effort to achieve stable autotrophic nitrogen removal (See Figure 1, Figure 2 and Figure 3). Startup procedures and reactor specifics were outlined in the previous reporting period.

At the current time, the reactor has been in operation for over 243 days. Results included in our previous report indicated that nitrogen removal was well below that which was expected in the literature for such a reactor configuration (Pynaert et al., 2004). As such, one of our main targets during this reporting period was to investigate operational techniques that would translate to increased nitrogen removal and improved reactor performance. Results included in this report consider December 25, 2004 to be day zero, and will focus most heavily on the recent performance period.

Sequential batch reactor (SBR) Construction/Operation.
Based on experience with the FBBR configuration, supplementary literature review, and recent global anammox successes with suspended culture systems, it was decided that implementation of a suspended culture configuration may be beneficial for project objectives. Use of a high density suspended growth reactor that can withstand periodic feeding (SBR) versus a high density biofilm reactor (FBBR) gives a broader range of comparison, and probably a more practical reactor design that will apply more readily to the field.

In the previous report, it was reported that a suspended growth continuous flow stirred tank membrane reactor (CSTMR) was being considered. One drawback with this configuration was the time delay in receiving membranes suitable for the flux required for a lab-scale system. To avoid project delay, a sequential batch reactor (SBR) configuration was subsequently chosen. Startup procedures and operational guidelines were then developed from existing OLAND and anammox literature. One of the options entailed running the reactor as a purely anaerobic system while the other option required implementation of an OLAND system through the use of intermittent aeration controlled by pH (Strous et al., 1999; Wett and Rauch, 2003). The latter option was initially chosen as the primary operational procedure. As will be discussed in the Preliminary Data section of this report, subsequent reactor performance dictated that operational parameters be changed.

The lab-scale SBR consists of a 5 L fermentation glass reactor vessel that was inoculated on June 27th, 2005 with 2.65 L of sludge showing nitritation/anammox activity. This sludge was obtained from Dr. Bernhard Wett at the University of Innsbruck, Austria. An aliquot (350 mL) of feed solution is added at the beginning of each reaction cycle, resulting in a total reaction/working volume of 3 L. A water bath is used to maintain a temperature of 35°C (See Figure 4 and Figure 6). The vessel was initially operated as described in Table 1.

The feed solution to the SBR is comprised of reject water obtained from anaerobically digested or anaerobically digested and dewatered sludge from Christiansburg, Virginia Wastewater Treatment Plant (WWTP). This reject water is pre-filtered through two (2) coffee filters followed by final filtration through 1.5 mm glass fiber filter. In cases when the thickening process was not operated, the sludge was first centrifuged at 10,000 x g before being filtered, as previously described. The feed is supplemented with ammonium chloride and/or sodium nitrite, as needed. The influent and effluent are analyzed for pH, alkalinity, COD, NH₃-N, NO₂^--N, NO₃^--N. Both pH and ORP (after July 7th) were monitored continuously except during the settle and decant phases.

This reactor was initially operated using a pH control strategy by Wett et al. (1998) to achieve oxygen limited OLAND conditions. In such a setup, nitritation and anammox are the processes that control the pH to within predetermined setpoints. Once the pH rises above the HIGH setpoint, aeration can be provided, which increases the bulk dissolved oxygen (DO) concentration and induces the acid-forming nitritation reaction. Once the pH drops below the LOW setpoint, aeration can be shut off and an anammox phase can begin during which pH is allowed to rise. Our equipment does not allow automated control at both the high and low pH setpoint; therefore, the reactor was aerated for 10 minutes at the end of each feed cycle. Unfortunately, excessive CO₂ stripping occurred during aeration, probably because the bicarbonate concentration was high enough to block pH control using the control strategy suggested by Wett, 2005 (unpublished). In addition to CO₂ stripping, nitrite accumulation occurred as well; both responses posed significant obstacles to achieving stable autotrophic nitrogen removal because they are known inhibitors of the anammox process (nitrite and high pH). As a result, the OLAND-based system was switched to a strictly anaerobic system on August 16th, 2005. To accomplish this switch, the biomass was first extracted and washed in a trace element solution to remove residual nitrite that had accumulated during reactor operation. The SBR was retrofitted with solenoid valves and degassed with 95%N2:5%CO₂ every 4 hours for 20 minutes (See Figure 5). This gas mix was chosen to avoid excessive CO₂ stripping and concomitant pH increases.

During the anaerobic phase, the filtered reject H₂O was initially used without dilution and supplemented with NH₄Cl and NaNO₂. Since NO₂^- has easily been generated through microaerobic treatment of the reject water, this step was implemented to focus on developing a bacterial community capable of achieving the anammox metabolism, knowing that the microaerobic nitritation process could be re-implemented later. Initially, the supplementation was at stoichiometrically balanced initial concentrations of 1,000 mg/L each of NH₄⁺ -N and NO₂^- -N. Hydroxylamine was also introduced during this period at an initial concentration of 10 mg NH₂OH/L since the literature seems to suggest that pulses of hydroxylamine can catalyze the initiation of the anammox reaction, especially when subjected to nitrite inhibition (Strous et al., 1999). This operation was maintained for seven (7) days during which nitrite accumulation was observed once again. It appeared that trace amounts of O₂ were still present, which allowed for the aerobic ammonia oxidizing bacteria (AAOBs) to oxidize NH₃ to NO₂^-. Phosphate concentrations in the influent were also above 50 mg P/L, which is listed as a threshold for complete inhibition of anammox type metabolisms (Strous et al., 1999). On August 27th, the biomass was extracted once again and washed with trace element solution to remove residual nitrite. To reduce the bulk DO concentration even further, the frequency of degassing was increased to every 3 hours for 20 minutes. Additionally, the influent feed composition was modified to achieve nitrite limitation during the reaction period by reducing the NO₂^--N and NH₄⁺-N concentrations in the feed to 100 and 500 mg/L, respectively, and supplementing with trace elements listed in Kuai et al., 1998. Hydroxylamine was dosed at 20 mg NH₂OH/L at the beginning of each reaction cycle. The reaction cycle time was increased to 24 hours. Finally, a constant influent flow was implemented to keep the ORP of the reactor within anoxic regions.

Rotating biological contactor (RBC) Construction/Operation.
After personal communication with Dr. Verstraete of the University of Ghent in Beligum, it was decided that an RBC may indeed be a viable reactor configuration for implementation of an OLAND based treatment system, as initially proposed. Literature review also provides evidence that dynamic conditions unique to an RBC may be beneficial to the syntrophic relationship between AAOBs and anaerobic ammonium oxidizers (AnAOBs) (Pynaert et al., 2003; Pynaert et al., 2004). Therefore construction of a lab-scale RBC is currently underway. The reactor has a total volume of 106 L. The reactor will be operated to ensure that 85% submersion of disk surface is achieved. Microaerobic conditions required for optimal OLAND conditions will be provided by gentle rotation of all disks along a horizontal shaft between 1 to 2.5 rpm. The influent flow rate will be maintained at 70.7 L/day with a hydraulic residence time of 1.5 days (See Figure 7 and Figure 8). Online pH and DO control will also be provided to maintain optimum growth conditions for both AAOBs and AnAOBs. Analysis of constituents similar to that performed for the FBBR will be carried out. At the same time, Wastewater Technology Inc, a manufacturer of full-scale RBC systems, is constructing a pilot-scale unit for use on the Virginia Tech dairy farm during year two. The unit will be complete and ready to use by the beginning of October 2005.

Based on the research team’s experience with anammox and OLAND systems over the past year, it is believed that evaluating reactor configurations for such systems are a critical piece of the full-scale implementation puzzle. Dr. Verstraete’s group at the University of Ghent has successfully maintained a lab-scale OLAND RBC for over 5 years, indicating that stable performance can be achieved and sustained. The stability of the OLAND process in an RBC configuration may indicate that the bacteria responsible for the OLAND metabolism thrive under dynamic conditions. In fact, the presence of AnAOBs in wastewater treatment systems was first discovered in an RBC reactor in Germany. Such dynamic conditions may select for syntrophic relationships that are currently being sought. The pilot-scale system will be implemented in the field with pre-processed waste from the dairy barn, while the lab scale system will be operated using synthetic wastewater as with the FBBR in order to test out operational strategies that may be useful at the pilot-scale.

Anaerobic Storage Unit Construction and Operation (Objective 1) and Winyah Bay Dairy Farm Nitrogen Balance (Objective 5).
As reported in the last report, a barrel-based anaerobic storage unit has been set up at the Virginia Tech dairy and receives flushed dairy manure from the operating Virginia Tech dairy barn. As shown in Figure 9, the pilot-scale system has started to form a crust on top of the waste, as expected in a conventional anaerobic stabilization system.

Three collaborator farms have been visited in the Winyah Bay watershed, and manure samples and data were collected. Wastewater composition for the anaerobically stabilized wastes from the Virginia Tech pilot scale system and the three Winyah Bay watershed farms is presented in Table 2. Additional information about the three farms visited is listed in Table 3. Significant variation was observed between farms for all analytes. The liquid from farm 1 is much higher in solids, COD total P and TKN than the other two farms, likely because of differences in waste handling. On farm 1, there is no attempt made at solids separation (as there is on farm 2) and the retention time in the lagoon is short (relative to farm 3), so less settling is possible.

Of these three farms, we have selected farm 3 to serve as the official collaborator for the study because the records available are more complete than the other farms. The anaerobic waste from this farm is intermediate in composition to the others, and its waste handling system (liquid waste accumulated and stored for ~3 months before land application) is more standard. The waste composition from this farm is similar to the anaerobically stabilized liquid from the pilot-scale anaerobic storage system being operated at Virginia Tech, which is consistent with proposal reviewer wishes for this project.

Information will be shared with all farms about the project, but we will focus on farm 3 for the whole farm nitrogen balance. A follow up visit is planned for late winter to collect remaining records needed for the nitrogen balance.

Difficulties
The planctomycetales bacteria responsible for anammox are extremely slow growing, with doubling times reported from 11 to 30 days (Van de Graf et al., 1996; Jetten et al., 2001). The process itself is also more than seven times slower than aerobic ammonia oxidation (Strous et al., 1998). Although slow growth is beneficial in a full-scale system that wants to reduce sludge management costs, it is challenging when trying to start up a system. The anammox process is also inhibited by various chemicals, including nitrite, oxygen and phosphate (van de Graaf et al., 1996; Jetten et al., 2001). Oxygen concentrations as low as 0.064 mg/L (2 mM) and nitrite concentrations between 70 and 140 mg NO₂^- -N/L (5 - 10 mM) inhibit the anammox activity completely, but reversibly (Jetten et al., 2001).

Fux et al. (2004) recently reported doubling times of up to 28 days in fixed bed reactors operated under conditions similar to those in this report. Observations of nitrite limitation and inhibition throughout the biofilm closely parallel observations in this report. As such, it is important to note that doubling times for pilot/full scale plants will not approach the doubling time of 11 days but may reflect periods closer to one month. The initial low concentration of inoculum and continued low biomass production in the FBBR has contributed to the slow response to reactor loading increases. As a result, the loading and removal rates reported here are lower than those reported by Pynaert et al., 2004 and Fux et al., 2004. Results indicate that reactor performance significantly drops every time the biomass is exposed to higher mass loadings. The prolonged periods of nitrite accumulation may have result in localized nitrite inhibition, which spreads throughout the biofilm and causes drastic performance slumps and a competitive disadvantage for the anammox bacteria in the biofilm.

Data collected over 243 days of operation of the FBBR indicate that the stoichiometry occurring in the reactor approaches that reported in the literature for anammox systems, suggesting that the AnAOBs are present. While aerobic ammonia oxidation is essential for the OLAND system, it is most critical to establish sufficient biomass capable of anammox activity prior to allowing competition with ammonia oxidizers. Thus, operational changes that have been made are estimated to be sufficient for conditions optimal for AnAOBs. The next step is to allow sufficient time for reactor recovery after periodic increases in mass loading to the system.

Operation of the SBR to achieve OLAND conditions proved to be a complicated task. One of the main reasons attributed to the high level of difficulty is the scale of the reactor (See Figure 6). The short water depth enhanced carbon dioxide stripping, which significantly impaired pH control. The configuration is much more conducive to selection of AnAOBs that can achieve anammox. As indicated previously, our strategy has shifted to establishing a critical mass of AnAOBs before introducing microaerobic conditions to the reactor, since AAOBs are seemingly able to proliferate in a short period of time.

Fortunately, oxygen uptake rate experiments and nitrate uptake rate experiments (data not shown) indicate that the COD provided in the reject water feed to the SBR is not sufficiently readily biodegradable (rbCOD) to encourage significant denitrification. Despite this fact, nitrate monitoring has continued to ensure that soluble microbial products as well as cell lysis are not providing sufficient rbCOD for heterotrophic denitrifiers to outcompete AAOBs for oxygen and AnAOBs for nitrite.

Project Objectives for Next Reporting Period

Objectives
All five objectives, listed previously, will be addressed to varying degrees during the next reporting phase. In addition, a modification to the original objective list is being proposed in the event that the OLAND system proves to be incompatible with dairy waste.

Tasks to Meet Objectives
Specific tasks to be completed over the next reporting period are:

1. Complete construction and begin operation of both the lab-scale and pilot-scale RBC.
2. Continue operation of the lab-scale SBR and RBC with synthetic feed to obtain stable autotrophic nitrogen removal and learn more about the process, to inform the pilot-scale studies.
   1. Summarize the ease and speed of start-up of each reactor configuration and report outcomes.
   2. Finalize characterization of the preliminary phase for each reactor configuration and determine the most appropriate operational and control parameters for satisfactory operation by:
      1. Assess stoichiometric validity of the anammox/OLAND processes within each reactor configuration.
      2. Characterize and confirm the presence of specific bacterial populations through use of fluorescence in situ hybridization (FISH) (methods already established in our laboratory for another project and is simply used as a diagnostic tool here)
3. Begin operating and evaluate the performance of the pilot-scale anammox/OLAND treatment systems fed either anaerobically stabilized dairy waste or effluent from an enhanced biological phosphorus removal (EBPR) pilot plant operating as part of another CICEET project.
   1. Automate and maintain stable operation over prolonged periods and achieve optimal biomass doubling time, as detailed in the literature.
   2. Assess stoichiometric validity of the aerobic ammonia oxidation and anammox processes using biomass from the OLAND reactors through kinetic batch tests.
   3. Assess reactor performance by establishing volumetric removal rates of NH₃ and NO₂^- after exposure to different oxygen and ammonia surface loadings.
   4. Characterize bacterial population shifts by FISH that may occur due to exposure to stabilized dairy wastes.
4. Consider using a SHARON-like process, which would use organic fermentation products to fuel conventional denitrification from nitrite, formed via microaerobic treatment of dairy waste for nitrogen removal. Compare ease of operation to OLAND.
5. Evaluate the degree of stabilization achieved by the pilot-scale anaerobic storage unit over time to assess its ability to achieve sufficient stabilization in support of the OLAND process. Determine rbCOD on the anaerobically stabilized material. Compare this to the effluent of the EBPR process that is low in phosphate and, therefore, may be more conducive to anammox treatment.
6. Initiate development of loading curves to simplify design of the OLAND treatment systems for application to dairy waste management.
7. Complete the nitrogen balance on a dairy farm in the Winyah Bay watershed, and a “what if” evaluation on the impact of implementing OLAND on that dairy farm.

Work Plan for Next Reporting Period
To accomplish tasks outlined previously, we will add either a second full time graduate student to the project starting in January 2006 or a post-doctoral researcher, who is currently being recruited from Japan. Furthermore, a part time technician is assisting with pilot-plant operations and maintenance in association with the Virginia Tech Dairy. Dr. Smets is traveling to the United States in late October to consult on the challenges associated with this project (Dr. Smets is a co-PI and has run anammox systems previously).

Anticipated Success in Meeting Project Objectives
The slow growing nature of the planctomycetales bacteria responsible for the anammox process within the OLAND system may prove to be a limiting factor. Obviously, the speed at which anammox biomass can be obtained through growth defines the progress of all elements of this report. Several minor modifications are currently being put into place to establish optimum growth conditions for the anammox biomass. Dr. Verstraete has promised to provide another dose of biomass from the RBC’s in his laboratory in Belgium in early Fall. This culture will be used to reseed the lab- and pilot-scale reactors to enhance their performance. Furthermore, the PI is pursuing use of a SHARON-like system, that will use more conventional nitrogen removal metabolisms by using fermented dairy waste as a fueling source for heterotrophic denitrification. We have learned substantial amounts about pilot-scale fermentation from our other CICEET project (focused on P removal from dairy waste) and believe this process will be easily achieved in the pilot plant phase. Therefore, we are confident that we will achieve N removal of dairy waste using one of these technologies. If anammox fails to perform, it will be an important outcome of this study.

Preliminary Data
FFBR Results
The FBBR was continuously loaded at 141 mg N/Lmedia.day for 46 days (See Figure 10). Nitrogen removal efficiency approached 65 ­ 70 % (See Figure 11); however, significant biomass production was not observed. Effluent total suspended solids (TSS) and volatile suspended solids (VSS) remained sufficiently low, indicating that biomass washout was not a significant factor on reactor performance (data not shown). On day 109, a decision was made to increase the loading to the reactor by 25% to 180 mg N/Lmedia.day. Following this increase, nitrite accumulation was observed initially as expected but began to decrease (See Figure 10). Such nitrite accumulation provides adverse conditions for AnAOBs, resulting in an overall performance deterioration.

On day 151, a reactor perturbance resulted in a further increase in nitrite concentration. To reduce nitrite concentrations, the reactor was operated in batch mode. Continuous feed was stopped from day 151 to day 160. Over this period, the nitrite concentration in the bulk liquid decreased considerably to below 10 mg/L. On day 162, continuous flow was resumed and the reactor was continuously loaded at 180 mg N/Lmedia.day until day 183 when the loading was increased once again by 25% to 226 mg N/Lmedia.day. It was our hope that this increase in the amount of mass loaded into the system would help to elucidate biomass activity levels. This increase in nitrogen loading prompted a slight accumulation of nitrite (See Figure 10). On day 190, the reactor was placed in batch operation in an attempt to reduce the nitrite concentration. Once again, this batch operation was successful in reducing the nitrite concentration to below detection limits on day 214.

The stoichiometry for anammox metabolism is:

NH₄⁺ + 1.26 NO₂^- + 0.085 CO₂ + 0.02 H⁺ = N₂ + 0.017 C₅H₇O₂N + 0.24 NO₃^- + 1.95 H₂O

As shown in Figure 12.gif, the stoichiometric conversion of ammonia and nitrite during continuous operation was consistent with what is expected from anammox cultures. Consequently, we believe that anammox was occurring in this anaerobic reactor, but biomass yield remained slow.

Additionally, on day 214 influent flowrate was increased to 4 L/day based on personal communications with other researchers dealing with anammox fixed film systems. The internal recycle flowrate was maintained at 9 L/day. Such an increase in flow was a deliberate attempt to achieve more adhesion of biomass to packing material and to also washout NOBs and debris that might have accumulated in the FBBR. The latter goal was included since our focus was to achieve optimum anammox conditions as these bacteria have an extremely slow growth rate and could possibly be outcompeted by NOBs if trace amounts of oxygen are present in the bulk liquid. Degassing of the recycle and feed jars with 95% N2:5%CO₂ every 4 hours for 10 minutes was also implemented on day 214 in an attempt to further reduce any possibility of bulk DO fluctuations. Implementation of this degassing system led to a decrease in the ORP values associated with the effluent of the FFBR (data not shown).

Between days 214 and 226, the reactor was continuously loaded at 226 mg N/Lmedia.day. On day 226, batch operation was once again implemented to allow the nitrite concentration to be reduced. Within 5 days of batch operation, nitrite concentrations were reduced to below detection limit. Additionally, ammonium concentrations were reduced to 6 mg N/L. Such successful reduction of nitrite concentrations suggested that utilizing such a configuration may increase overall N removal within the reactor. Thus it was decided that the reactor would be operated in a semi-continuous/batch mode to investigate the N removal potential of such an operational configuration.

In shifting operation procedures, there was the need to assign reaction times for each phase of operation. To accomplish this, a cycle analysis was performed to determine a protocol for reactor operation. The cycle consisted of continuous flow for 2 days followed by a batch reactor time of 3 days. From this cycle analyses, it was shown that the ammonia concentration in the reactor steadily decreased over the 5 days to a final value of below 1 mg N/L. At the same time, nitrite concentrations decreased from 20 mg N/L to 14 mg N/L and nitrate concentrations increased from 11 mg N/L to 19 mg N/L. Such N removal data initially seemed promising. To explain this improvement in removal, we considered the possibility that the dynamic conditions induced by semi-batch operation may have stimulated the organisms responsible for N removal. The exact mechanism for such stimulation of removal was not known but based on communications with Dr. Verstraete, we considered the possibility that nitrogen removal may have been enhanced due to possible production of the metabolic intermediate hydroxylamine, which contributed to the reduction of accumulated nitrite. Hydroxylamine is an intermediate of both anammox and nitrification. In aerobic ammonia oxidation, ammonia monooxygenase (AMO) converts ammonia to hydroxylamine before hydroxylamine oxidoreductase (HAO) further converts it to nitrite. In the anammox process, HAO can oxidize hydroxylamine but can also reduce nitrite and nitric oxide to nitrous oxide (Schalk et al., 2000). Hypothetical models for the anammox process have suggested that nitrogen dioxide helps to catalyze the oxidation of ammonia to hydroxylamine. Hydroxylamine then is reduced to nitrite as stated previously (Schmidt et al., 2002). Kuai et al. (1998) reported that simultaneous removal of ammonia and nitrite occurred once hydroxylamine was added to batch vessels, indicating that the anammox process could be stimulated with pulses of hydroxylamine. It was postulated that the batch mode could have stimulated existing AAOBs to produce trace amounts of hydroxylamine which was then utilized by AnAOBs.

Further examination of the data, however, indicated that stoichiometric trends for the semi-batch mode of operation varied greatly from the expected anammox stoichiometry (see Figure 13). The batch operation trends were different from the stoichiometry observed during continuous flow operation, which approached reported anammox stoichiometry (See Figure 12). There appeared to be excessive production of nitrate once the semi-batch mode was implemented, suggesting that NOB activity was stimulated. Such stimulation could have occurred when headspace increased in the recycle jar due to frequent sampling. This increase in headspace may have provided a source of oxygen via entrained air, and allowed the NOBs to outcompete the AnAOBs. Additionally, the dynamic operation (continuous flow to batch to continuous flow) may have an impact on the ability of AnAOBs to remain active in the system, thereby resulting in anomalous stoichiometry. Results from this cycle also indicated that further optimization of the scheduled semi-continuous/batch operation is needed if it is to be implemented on a permanent basis.

At this time, we have decided to return to a continuous flow operational configuration. We hope to re-establish conditions most suitable for AnAOBs to outcompete NOBs. Heavy monitoring of reactor nitrite and nitrate concentrations is currently taking place. Results from these analyses will dictate the daily operational procedure implemented. If nitrite accumulation is observed, the continuous feed will be discontinued and the reactor placed into a batch mode. However, once nitrite concentrations are reduced below inhibitory levels, continuous flow will be reestablished. Based on current and past results, the investigators believe that such operation is the optimal way towards improving nitrogen removal efficiency.

SBR Results
The SBR has not operated well since the first few days after initiation (Figure 14). It is believed that oxygen entrainment is a problem. The system was originally operated to achieve both aerobic nitritation (conversion of ammonia to nitrite) and anaerobic ammonia oxidation (ammonia removal using nitrite from nitritation process as electron acceptor and ammonia as electron donor). However, this proved to be difficult to control in a short reactor due to oxygen transfer issues; consequently, the system has been converted to an anerobic system only in an attempt to sustain AnAOBs. At this time, the system is being tightly monitored to evaluate its performance. Hydroxylamine is also provided as a supplement to stimulate anammox metabolism.

RBC Results
Construction of the lab-scale and pilot-scale RBCs is underway. Startup and maintenance will commence by mid- September. Results from this operation will be included in the next reporting period.

Dissemination
Publications:
None yet

Workshops:
None

Conferences:
W. Khunjar, P. Sweetman, N. G. Love, K. F. Knowlton, B. F. Smets. Treatment of Anaerobically Stabilized Dairy Waste with an Oxygen Limited Autotrophic Nitrification plus Denitrification (OLAND) Fixed Film Reactor: Startup and Maintenance Issues. Virginia VWEA/AWWA Water Jam Conference, September 29, 2005.

Manuals, Protocols:
None

Outreach Activities:
None

Contact with End Users:
The PIs have communicated with the primary end-user, Dale Gardner of the Virginia State Dairyman’s Association regarding our progress. His feedback is being obtained by letter, to be submitted soon.

Patent, Copyright, Invention Disclosure Activity:
None at this time. If the process is shown to function for dairy waste, there may be intellectual property to be disclosed at that time.

Expenditures
The project is 50% complete based on time and only 32% complete based on expenditures. The low rate of expenditure reflects the fact that the major bioreactor construction costs have not hit the budget reports yet, and the primary student involved with this project is on a fellowship, so his stipend does not cost as much as budgeted. As noted above, we are in the process of adding either a second graduate student or a post-doctoral research associate to the project to assist the current student, which will utilize the savings realized due to his fellowship, plus will provide the additional manpower needed to achieve the goals of year 2.

End User Advisor
Name: Dale Gardner, Executive Secretary
Organization: Virginia State Dairyman’s Association
Location: Harrisonburg, VA
Phone Number: 540-434-6722
Email: vamilk4u@gte.net

References
Fux, C., Marchesi, V., Brunner, I., Siegrist, H. Anaerobic ammonium oxidation of ammonium- rich waste streams in fixed-bed reactors. Wat. Sci. Tech 2004; 49/11: 77-82.

Jetten MSM, Wagner M, Fuerst J, van Loosdrecht M, Kuenen G, Strous M. Microbiology and application of the anaerobic ammonium oxidation (‘ANAMMOX’) process. Curr Opin Biotechnol 2001;12: 283­ 8.

Kuai L, Verstraete W. Ammonium removal by the oxygen-limited autotrophic nitrification­denitrification system. Appl Environ Microbiol 1998;64: 4500­ 6.

Pynaert, K., Smets, B.F., Beheydt, D., Verstraete, W. Startup of Autotrophic Nitrogen Removal via sequential biocatalyst addition. Environ. Sci. Technol. 2004; 38: 1228-1235.

Pynaert, K., Smets, B.F., Beheydt, D., Verstraete, W. Characterization of an Autotrophic Nitrogen-Removing Biofilm from a Highly Loaded Lab-Scale Rotating Biological Contactor. Appl Environ Microbiol. 2003; 69: 3626-3635.

Schalk, J., Vries, S. de, Kuenen, J.G., Jetten, M.S.M. Involvement of a Novel Hydroxylamine Oxidoreductase in Anaerobic Ammonium Oxidation. Biochemistry. 2000; 39: 5405 ­ 5412.

Schmidt, I., Sliekers, O., Schmid, M., Cirpus, I., Strous, M., Bock, E., Kuenen, J.G., Jetten, M.S.M. Aerobic and anaerobic ammonia oxidizing bacteria ­ competitors or natural partners? FEMS Microbiology Ecology. 2002; 39: 175 ­ 181.

Strous M, Heijnen JJ, Kuenen JG, Jetten MSM. The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microorganisms. Appl Microbiol Biotechnol 1998; 50: 589­96.

Strous, M., Kuenen, J.G., Jetten, M.S.M. Key Physiology of Anaerobic Ammonium Oxidation. Appl Environ Microbiol. 1999; 65/7: 3248-3250.

van de Graaf AA, de Bruijn P, Robertson LA, Jetten MSM, Kuenen JG. Autotrophic growth of anaerobic ammonium oxidizing microorganisms in a fluidized bed reactor. Microbiology 1996;142: 2187 ­ 96.

Wett,B. Solved scaling problems for implementing deammonification of rejection water. Unpublished 2005.

Wett, B., Rostek, R., Rauch, W., Ingerle, K. pH-controlled reject water treatment. Wat. Sci.Tech 1998; 37/12: 165-172.

Wett, B. and Rauch, W. The role of inorganic carbon limitation in biological nitrogen removal of extremely ammonia concentrated wastewater. Water Research 2003; 37/5: 1100-1110.